It should be possible to design a Carbon Border Adjustment Mechanism (CBAM) that is World Trade Organization (WTO) compatible, but this depends upon the whole design of the regime and how it is implemented. Basing a CBAM on an emissions trading system was not a way favored by participants – although some thought it could be made WTO compatible. There was considerable support for a regime based on a carbon excise tax to be charged on certain high carbon intensity products; such a tax could be equally charged on imports and be refunded on export. It could be linked with a subsidy or tax rebate regime to encourage greater emissions reduction (i.e. companies with higher than average emissions reduction would be partially compensated)

Background to the Carbon Border Adjustment Mechanism (CBAM)

As of 2020, 189 countries have signed the Paris Agreement agreeing to limit their CO2 emissions. However, the Paris Agreement permits countries to set their own ambitions -within certain parameters. Some jurisdictions or regions have undertaken to cut carbon emissions faster than others. The EU, in particular, has stated its ambition to cut emissions by 2030 by 55% in comparison with 1990 levels1. However, increasing ambitions raises concerns about "carbon leakage". This is the concern that carbon emissions might be simply shifted to low-ambition jurisdictions, as producers move activity to avoid paying a carbon price, or as consumers in a high ambition jurisdiction will be encouraged to purchase imports produced in low ambition jurisdictions where the cost of production is lower. The consequence is that carbon leakage reduces the impact of carbon reduction measures in a high ambition jurisdiction. It may lead to job losses and lack of competitiveness in affected industries in the high ambition location.

While, to date, significant carbon leakage has not been observed in empirical studies, this may because carbon prices are relatively low and regimes such as in the EU already have mechanisms to reduce leakage (the EU grants free allowances under the EU Emissions Trading Scheme to reduce leakage.) The concern is that increasing emissions reduction ambitions, and consequently the carbon price, will drive actual leakage.

In order to address the environmental concerns related to carbon leakage, the EU Commission released a consultation in 2020 on the possibilities to introduce a carbon border adjustment mechanism (CBAM). This could take a number of forms. It could be a tax on imports, a mechanism based on the EU Emissions Trading System (ETS), or a new tax charged both within the EU and on imports based on the average carbon intensity of certain products. If the CBAM is based on the EU ETS it is possible that it would simply be a type of tax on imports charged according to the prevailing price of ETS permits; alternatively it might be a shadow ETS system under which an importer would buy permits, again at a price based on the prevailing EU ETS price.

The possibility of jurisdictions implementing a CBAM have been discussed by environmental experts for many years, but it now seems that there may be genuine political will to implement something. While such measures have been raised in other jurisdictions, the EU appears to be furthest ahead with proposals. With this in mind, KPMG IMPACT2  has decided to hold a number of roundtables, as part of the Global Responsible Tax Project, looking at how a CBAM could be crafted, the implications for World Trade Organization (WTO) compliance and practical issues for corporates.


The first roundtable was held with a group of experts on 9 March 20213. The discussions were focused on a CBAM in general. They were not intended solely as a critique of any EU Commission proposals.

Attendees spoke in their private capacity rather than on behalf of any organization. The meeting was held under the Chatham House rule and any attributed comments have been cleared with the speaker.

In some cases there was general agreement among the participants but on other points there was disagreement. The views expressed are not intended as advice and are not necessarily the views of any KPMG firm.

Outline of WTO issues

It should be possible to design a CBAM around a carbon tax which would be WTO compliant. Potentially the same is true if it is based on an ETS although this raises further complexities. It is also likely that a CBAM based on either approach would have to rely on the General Exceptions in the General Agreement on Trade and Tariffs which would mean its purpose would have to be exclusively environmental and not economic protection. If competition considerations were to be privileged in developing and devising the CBAM, no deviation from the national treatment and most favoured nation principles4 could be justified under the General Exceptions.**

The roundtable began with a brief overview of some key WTO provisions with which any CBAM would need to comply.

While there is a general prohibition on a jurisdiction applying tariffs on imports (other than in accordance with the general tariff schedule) there is an exception for charges which are equivalent to internal taxes, provided they are not charged in a discriminatory way on imports (General Agreement on Trade and Tariffs, Article II.2.a). This means that it should be possible to make a carbon border tax WTO compliant provided it was equivalent to a domestic carbon tax.

However, there are a number of issues which need to be resolved. First a tax may only be adjusted at the border if it is an indirect tax on a product and not a direct tax on a producer. One area of dispute could be whether a carbon tax is actually a direct or indirect tax (although it is generally considered to be an indirect one). WTO jurisprudence also establishes that a tax can be adjusted if it is on an input that is physically present in the final product but it is unclear if this also applies to a by-product of the production process - like carbon emissions - which are not physically present in the final product.

If these issues are not resolved it is possible that there could be a WTO challenge of a CBAM. One way to resolve this would be for there to be an agreement on the interpretation of a carbon tax which would require consent of three quarters of the members of the WTO.

Subject to these interpretation issues, it should be possible for a CBAM based on a carbon tax to be WTO compliant. As discussed below, some more complex issues would arise if the CBAM was based on an ETS.

Nevertheless, even if there is a prima facie breach of WTO rules, a CBAM may be saved by the General Exceptions. These allow, for example, rules to be introduced to protect health or exhaustible natural resources. However, it is important to note that this exception would only apply if the purpose of the CBAM was exclusively to prevent the environmental effects of carbon leakage. Often in discussions about a CBAM, the need to protect the competitiveness of domestic industry is raised. Any implication that the CBAM was, in part, needed for economic purposes would prevent the application of the exception.

General comments on the structure of a CBAM

A key issue for WTO compliance will likely be whether importers are subject to discrimination, including through administrative burden or the practical operation of a CBAM. If the CBAM was to be based on an ETS it should not be relevant whether it is a shadow ETS regime or some kind of import tax calculated accorded to the prevailing ETS permit price

There was general agreement that the ideal system would be some kind of globally agreed carbon tax – and it was suggested that could be on the principle of common but differentiated responsibilities, with countries establishing an internationally agreed price at the domestic level in accordance with e.g. their contributions to global emissions, their GDP, etc. But it was recognized that this was unlikely to happen (at least in the short to medium term). Any carbon pricing mechanism and CBAM should be as administratively simple as possible and give producers predictability and certainty over the carbon price. It would also helpful for governments to use available private sector instruments, such as environmental certifications.

There was general agreement that whether or not any form of CBAM would be compliant with WTO rules was a complex issue and would depend upon the package as a whole. This also involves looking at the compliance burden, the way the system was applied in practice, and the extent of discussions with other WTO members to try and ensure any regime was as least disruptive to trade as possible.

WTO rules apply a substance over form approach and therefore the structure to achieve the CBAM would be less important than the actual effect and whether or not importers were subject to discrimination. For this reason it was thought that it was not relevant whether the EU decided to introduce a CBAM as a shadow ETS system - whereby importers would actually have to purchase permits - or whether the system was introduced as some kind of tariff or import duty based on the price of EU ETS permits.

It was also agreed that it would be possible to apply the CBAM to certain products and not others - provided there was no discrimination between like products. Therefore, if a jurisdiction wished to introduce a CBAM, it could start by imposing it on specific products such as steel, aluminum and cement and then extend the system to other products at a later date if desired.

Considerations about a CBAM based on an ETS

There was a disagreement over whether a CBAM based on an ETS would require the immediate abolition of free allowances under the domestic ETS or whether it would be possible to take account of the impact of free allowances in calculating cost to importers so as to avoid any discrimination. Providing a rebate to support exporters would breach WTO rules

There was some disagreement over whether providing free allowances under the existing EU ETS is already in breach of WTO rules – although it has never been challenged. Currently, in order to mitigate against carbon leakage, some ETS permits are issued for free. It was suggested that providing free allowances was an illegal subsidy under WTO rules. However, this view was contested. It was pointed out that the free allowances are issued on the basis that they would cover carbon emissions by a plant using the lowest emissions technology available. In practice this is never the case (at least as regards an example discussed about the steel industry) which meant that all producers have to purchase some of the allowances – which therefore puts a price on carbon.

There was a discussion about whether or not it would be possible for the EU to continue with granting free allowances to EU producers if a CBAM was introduced. The concern would be that there would be discrimination against importers who would not benefit from free allowances as they would not actually be part of the EU ETS. Nevertheless, the contrary view is that the charge on importers could be structured to take account of free allowances they would have received had they been located within the EU. Therefore, the carbon price they would incur would effectively be the same as for an EU producer and they would not actually be subject to discrimination.

It was also noted that the EU has indicated that the free allowances will be phased out by around 2030. Concerns have been expressed, for example in the European Parliament, that phasing out such allowances will adversely affect exports and therefore jobs. Nevertheless, if the EU CBAM is based on the EU ETS it would be very difficult to provide support for exports. This is because any rebate on exports would have the goal of supporting the competitiveness of EU industry rather than preventing carbon leakage as such. It would therefore undermine the argument that the CBAM was entirely an environmental instrument which would prejudice the application of the General Exceptions in the WTO rules.

The potential to structure a carbon tax as a consumption or excise tax

If a domestic carbon tax was applied like a consumption or excise tax it would be possible to apply it at the border on imports and comply with WTO rules. It would also be refundable on exports – as it would only be a charge on domestic consumption – and so would support exporters

There was considerable support for basing carbon pricing and a CBAM on a consumption-based tax similar to the way in which tobacco or alcohol is taxed in many countries. This would involve charging a tax on consumption of products such as steel, glass, and cement. The tax could be differentiated per product according to the average carbon intensity of that type of product. Such a tax could be charged on imports as it would not be discriminatory - it would apply to the same product irrespective of whether it was made inside or outside the jurisdiction. The tax could also be refunded on exports as it would only be applied on the consumption of the product within the domestic jurisdiction. It would therefore operate in the same way as many excise duties.

The advantage of structuring the tax in this way is that it could be made WTO compliant, but it would also protect exports as, by its nature, it would be refunded on export.

There was some debate about the extent to which such an excise type tax would prevent carbon leakage. As it would be based on the average carbon intensity of a product (not the actual intensity of a particular product) it is possible that importers from low ambition jurisdictions would still have a competitive advantage. The difference between the average carbon intensity and the actual intensity of the product would effectively not be subject to a tax on import or a carbon price.

As a carbon excise tax would be based on the average carbon intensity of a product it would be necessary to couple it with a subsidy or rebate mechanism to encourage and reward greater emission reduction efforts

It was noted that the excise tax approach would not encourage domestic producers to reduce their carbon footprint. Therefore, for environmental reasons, it would be necessary to also have some kind of subsidy or rebate system which would incentivize low carbon production. This subsidy or rebate would need to apply equally to imports to avoid discrimination and possible breach of the WTO rules.

It was agreed subsidies do not necessarily breach the Agreement on Subsidies and Countervailing Measures; in particular it is important that they do not have an adverse effect on the marketplace.

There was no conclusion on how the subsidy or rebate mechanism would actually work but it was pointed out that in many industries there are green certification programs which could be used in order to determine the extent to which a subsidy was due on a particular product. It was also thought that the mechanism should be structured as a subsidy rather than a rebate of the carbon excise tax as varying the tax according to emissions intensity could raise WTO complexities.

There was some discussion as to whether the carbon excise tax should be set according to the global average carbon intensity of a product or on the EU average. It was noted that if it was based on the EU average this should reduce the number of companies applying for a subsidy - on the basis of applying more efficient carbon reduction methods than the average used to set the tax - which would reduce complexity in the system.

International credits or Carbon Offsetting

There was some support for allowing international credits (or carbon offset mechanisms) within a CBAM**

One participant noted the importance of allowing carbon offsetting - which allows companies to reduce their net emissions in the most cost-effective way. International credits (akin to carbon offsets) are, however, no longer eligible for credit under the EU ETS. It was suggested that, ideally, any CBAM should incorporate such a mechanism given the global nature of carbon emissions

The need for a carbon club and international cooperation

There was a general belief that international cooperation to achieve effective minimum carbon pricing was more important than implementing a CBAM

There was broad support for some kind of carbon club or carbon customs union. One participant noted that the EU has set a very high carbon reduction ambition and this would best be achieved through dialogue with other jurisdictions and regions rather than through a CBAM. Such a mechanism could be used to move towards a minimum global carbon price. It was agreed that a carbon club could be WTO compliant – unless the members implemented rules which discriminated against non-members. In that regard, any measures would be judged in the same way as if they were imposed unilaterally by one country.

It was pointed out that the World Bank has already formed a Coalition of Finance Ministers for Climate Action, to discuss carbon pricing.

Such international cooperation may also assist in developing the principle of common but differentiated responsibilities which would enable developing countries to take measures to increase carbon prices proportionately while taking account of their specific needs to protect domestic industry.

There was also a suggestion that WTO rules could be amended to allow for a CBAM – although it was recognized that may not be politically feasible.